Cigarette smoking is associated with increased risk for several chronic diseases including chronic obstructive lung disease. Traditional smoking cessation programs have generally produced disappointing long-term results and have failed to attract a substantial percentage of smokers. As a result, investigators have begun exploring alternative treatment approaches and delivery systems. Controlled smoking, originally developed by Frederiksen, is one such alternative approach which focuses upon making reductions in the smoking response dimensions of substance, rate, and topography. As employed in recent studies in our laboratory, controlled smoking can be offered to both individuals wishing to stop smoking and to those wishing to reduce but not eliminate their smoking. The current proposal will evaluate the long-term effectiveness of controlled smoking when offered in worksite settings and attempt to identify predictors of treatment outcome. An initial study will evaluate the incremental effectiveness of adding social support (a buddy system) and competition/incentive procedures to a basic controlled smoking program. Dependent variables will include self-monitoring records of smoking rate, brand, and percent of the cigarette smoked as well as carbon monoxide and saliva thiocyanate levels. A second study will assess the comparative effectiveness of controlled smoking relative to an alternative dosage reduction procedure and to a multi-component cessation based program. A third study will evaluate several alternative delivery systems for controlled smoking including group vs. individually administered treatment, a self-help manual, and a computer assisted program. In summary, this proposal focuses on conducting controlled evaluations of an innovative and broadly applicable smoking control program. If this approach proves successful and such programs can be offered in cost-effective formats in worksite settings, they can make a significant contribution towards health promotion and disease prevention.